In recent years, with the widespread use of cordless and portable AV appliances, personal computers and the like, the need has been increasing for compact, light weight, and high energy density batteries as power sources for driving those appliances. In particular, lithium secondary batteries, as having high energy density, are expected to be dominant batteries in the next generation, and the potential market thereof is very large.
In most of the lithium secondary batteries currently available on the market, LiCoO2 having a high voltage of 4 V is used as the positive electrode active material, but LiCoO2 is costly because Co is expensive. Under such circumstances, research has been progressing to investigate various positive electrode active materials as substitutes for LiCoO2. Among them, a lithium-containing transition metal oxide has been wholeheartedly researched: LiNiaCObO2 (a+b≈1) is promising, and it seems that LiMn2O4 having a spinel structure has already been commercialized.
In addition, nickel and manganese as substitute materials for expensive cobalt have also been under vigorous research.
LiNiO2 having a layered structure, for example, is expected to have a large discharge capacity, but the crystal structure of LiNiO2 changes during charging/discharging, causing a great deal of deterioration thereof. In view of this, it is proposed to add to LiNiO2 an element that can stabilize the crystal structure during charging/discharging and thus prevent the deterioration. As the additional element, specifically, there are exemplified cobalt, manganese, titanium and aluminum.
Moreover, prior art examples which use composite oxides of Ni and Mn as the positive electrode active material for lithium secondary batteries will be described: U.S. Pat. No. 5,393,622, for example, proposes a method in which a hydroxide of Ni, a hydroxide of Mn and a hydroxide of Li are dry-mixed together and baked and, after cooling them down to room temperature, the mixture is again heated and baked to obtain an active material having a composition represented by the formula LiyNi1-xMnxO2 where 0≦x≦0.3, 0≦y≦1.3.
Further, U.S. Pat. No. 5,370,948 proposes a method in which a Li salt, a Ni salt and a Mn salt are mixed together into an aqueous solution, followed by drying and baking, to obtain an active material represented by the formula LiNi1-xMnxO2 where 0.005≦x≦0.45.
Further, U.S. Pat. No. 5,264,201 proposes a dry synthesis method in which hydroxides or oxides of nickel and manganese and an excess amount of lithium hydroxide are mixed together and baked, and a synthesis method in which an oxides of nickel and manganese or the like are added to a saturated aqueous solution of lithium hydroxide to form a slurry, which is then dried and baked under a reduced pressure, to obtain an active material represented by the formula LixNi2-x-yMnyO2 where 0.8≦x≦1.0, y≦0.2.
Furthermore, U.S. Pat. No. 5,629,110 proposes a dry mixing synthesis method which uses β-Ni(OH)2 to obtain an active material represented by the formula LiNi1-xMnxO2 where 0<x≦0.2, y≦0.2.
Japanese Laid-Open Patent Publication No. Hei 8-171910 proposes a method in which manganese and nickel are coprecipitated by adding an alkaline solution into an aqueous solution mixture of manganese and nickel, then lithium hydroxide is added to the aqueous solution mixture and the resulting mixture is baked to obtain an active material represented by the formula LiNixMn1-xO2 where 0.7≦x≦0.95.
Further, Japanese Laid-Open Patent Publication No. Hei 9-129230 discloses a preferable particulate active material having a composition represented by the formula LiNixM1-xO2 where M is at least one of Co, Mn, Cr, Fe, V and Al, 1>x≧0.5, and shows a material with x=0.15 as the active material containing Ni and Mn.
Further, Japanese Laid-Open Patent Publication No. Hei 10-69910 proposes an active material synthesized by a coprecipitation synthesis method, represented by the formula Liy-x1Ni1-x2MxO2 where M is Co, Al, Mg, Fe, Mg or Mn, 0<x2≦0.5, 0≦x1<0.2, x=x1+x2, and 0.9≦y≦1.3. This patent publication describes that the discharge capacity is inherently small if M is Mn, and the essential function of the positive electrode active material for a lithium secondary battery intended to achieve a high capacity is dismissed if x2 is more than 0.5. LiNi0.6Mn0.4O2 is exemplified as a material having the highest proportion of Mn.
It should be noted that, although U.S. Pat. No. 5,985,237 shows a production method of LiMnO2 having a layered structure, this is practically a 3 V level active material.
All of the prior art examples disclosed in the above U.S. Patents and Japanese Laid-Open Patent Publications are intended to improve the electrochemical characteristics such as the cycle characteristic of LiNiO2 by adding a trace amount of an element to LiNiO2, while retaining the characteristic properties of LINiO2. Accordingly, in the active material obtained after the addition, the amount of Ni is always larger than that of Mn, and the preferable proportion is considered to be Ni:Mn=0.8:0.2. As an example of a material having a proportion with a highest amount of Mn, Ni:Mn=0.55:0.45 is disclosed.
However, in any of these prior art examples, it is difficult to obtain a composite oxide having a single-phase crystal structure since LiNiO2 is separated from LiMnO2. This is because Mn2+ is likely to be oxidized to Mn3+ during coprecipitation, and Mn3+ is unlikely to form a homogenous composite oxide with Ni2+.
As described above, as a substitute material for the currently commercialized LiCoO2 having a high voltage of 4 V, LiNiO2 and LiMnO2 as high-capacity and low-cost positive electrode active materials having a layered structure like LiCoO2 have been researched and developed.
However, the discharge curve of LiNiO2 is not flat, and the cycle life is short. In addition, the heat resistance is low, and hence the use of this material as the substitute material for LiCoO2 would involve a serious problem. In view of this, improvements have been attempted by adding various elements to LiNiO2, but satisfactory results have not been obtained yet. Further, since a voltage of only 3 V can be obtained with LiMnO2, LiMn2O4 which does not have a layered structure but has a spinel structure with low-capacity is beginning to be researched.
Namely, required has been a positive electrode active material which has a voltage of 4V, as high as LiCoO2, exhibits a flat discharge curve, and whose capacity is higher and cost is lower than LiCoO2; further required has been a high-capacity non-aqueous electrolyte secondary battery with excellent charge/discharge efficiency, which uses the above positive electrode active material.
As opposed to this, Japanese Patent Application No. 2000-227858 does not propose a technique for improving the inherent characteristics of LiNiO2 or those of LiMnO2 by adding a new element thereto, but proposes a technique for obtaining a nickel manganese composite oxide (positive electrode active material) which represents a new function by dispersing a nickel compound and a manganese compound uniformly at the atomic level to form a solid solution.
That is to say, the prior art examples propose plenty of additional elements, but not technically clarify which elements are specifically preferred, whereas the above application proposes the positive electrode active material which can represent a new function by combining nickel and manganese at about the same ratio.
Based on the fact that it is possible to obtain a nickel manganese composite oxide which represents a new function by dispersing nickel and manganese uniformly at the atomic level to form a solid solution, the present inventors conducted extensive studies on oxides containing a variety of transition metals, the composition and the crystal structures thereof.
Therefore, it is an object of the present invention to provide a positive electrode active material comprising a composite oxide, which represents a further new function, by means of a technique for forming a solid solution by uniformly dispersing three transition metal elements, instead of two transition metal elements, at the atomic level.